# Shopify E-Commerce Analytics and Decision Memory: What Every Tool Captures Without Being Asked > Every tool in this category publishes what it costs. None of them publish, on that pricing page, the variable that matters most: whether the tool fills itself or waits for someone to do the work. Source: https://saascode.ai/inside/shopify-e-commerce-analytics-and-decision-memory-what-every-tool-captu · Published: 2026-08-29 · Section: listicle · Product: Shopgraph (https://saascode.ai/products/shopgraph) --- Every tool in this category publishes what it costs. None of them publish, on that pricing page, the variable that matters most: whether the tool fills itself or waits for someone to do the work. That question — what does this tool capture without being asked? — is the one that decides whether a decision-memory product is useful after twelve months or empty from the first billing cycle. It is also the criterion pricing pages skip, because it is harder to promise and harder to verify than a feature list. **Shopgraph** is a codebase you purchase once and deploy as a multi-tenant SaaS for Shopify sellers in the $500K–$10M GMV band. It captures business decisions automatically from Shopify data, binds them to what the numbers did next at seven and thirty days, and answers questions in prose over each seller's private corpus. This page maps the full landscape an operator would be entering: from the free platform-native floor through the indie-app tier through the analytics subscriptions and up to the own-the-code option. --- ## Why this is the axis pricing pages skip Shopify, which has more data on merchant behaviour than any other platform in this category, drew its own conclusion when it built its annotation system: it shipped automatic event markers in May 2026 specifically without a manual-entry path. Its help centre states it plainly — "You can't create or add your own annotations manually" — because the platform determined that context is valuable and the manual-logging part is what merchants will not reliably do. That product decision is the best available evidence in this guide, more reliable than any practitioner statistic, because it represents a well-resourced company's bet on which half of the adoption problem is real. Every tool on this list can be placed on a spectrum from "derives everything" to "requires the merchant to type everything." That placement decides whether the corpus is useful at six months or empty at twelve, and it is the one variable worth reading past the pricing pages to find. --- ## Five criteria that differentiate tools in this category What each tool captures automatically, and what it waits for. This is the guide's organising variable. Whether the seller can query their own annotated history in natural language, and what the corpus under that query contains. Some tools offer an AI assistant over live store data; others offer a query interface over a structured record of past decisions. The underlying corpus determines what kinds of questions the tool can answer. Whether the tool connects a specific decision to what followed it at a measured interval. A log without an outcome is a diary. Outcome binding — pulling post-event metrics at a set interval — is what turns a diary into evidence. How the tool's pricing scales with seller growth. Three tools in this market meter on annual GMV or media spend; a seller's bill rises as their revenue rises. An operator mapping this market needs to understand that model — both as a business fact about the businesses they are entering alongside or displacing, and as the question a seller will raise in a renewal conversation. Who controls the data and the platform. Most tools here are subscriptions the seller rents. One entry is a codebase. The ownership structure determines who keeps the margin, who holds the corpus, and what a seller's history is worth if the vendor gets acquired. --- ## The entries, grouped by what they capture without being asked ### Automatic, zero-work — the platform-native floor **Shopify analytics, annotations, Sidekick, and Rollouts** Price: $0 incremental, included on every Shopify plan. Sources: Shopify Changelog 2026-05-05 and 2026-07-29 (both PRIMARY); shopify.com/sidekick (PRIMARY). In May 2026, Shopify began drawing coloured event markers on every merchant's analytics chart automatically, based on store activity — product launches and removals, theme publications, app installs and removals, platform data delays. No setup required. In July 2026, Shopify extended the system to allow third-party apps to write their own markers covering marketing campaigns, discounts and promotions, inventory and fulfillment changes, and custom event categories. Alongside this, Shopify Sidekick is a natural-language AI assistant on every plan that answers questions over live store data and proactively surfaces observations on the admin home. Shopify Rollouts provides native A/B testing of themes and checkout at no additional cost. The platform's own help centre draws the line: "You can't create or add your own annotations manually." Annotations come from store activity or apps — a merchant cannot type their own decision. Sidekick's memory is conversational context over current data, not a structured corpus of annotated past decisions with retrospective outcomes. **Honest limit:** If a seller's questions are answered by what Shopify derives and displays automatically, the free tier is the correct recommendation. Naming this honestly to every prospect is what earns the credibility that makes the rest of the pitch work. --- ### Automatic plus manual — the indie app tier **Annotrack – Store Annotations** (Shopify App Store, developer: The Fresh Lab Ltd) Price: Free (up to 25 annotations) · Growth $9/mo · Unlimited $15/mo. Source: apps.shopify.com/annotrack-annotations, 2026-08-07, PRIMARY. Annotrack launched on the Shopify App Store on May 19, 2026. Its core: "Log campaigns, promotions, site changes, app installs, or any operational event with a date, category, and note." Paid tiers add impact reports — performance before and after each annotation — and integrations that pull Klaviyo and Mailchimp campaign events automatically. On capture method, Annotrack is a structured text box for manual entry with optional integrations layered on. The Klaviyo and Mailchimp connections provide the derived half, creating annotations from email campaign events without the seller typing. Its App Store listing showed no reviews approximately eleven weeks after launch — a supply-exists signal and a demand-unknown signal, simultaneously. **Honest limit:** Annotrack does the annotation job at an honest price for sellers who will use it. It does not provide semantic querying over the corpus, derived capture at the scale of a full Shopify connector, or a multi-tenant SaaS the operator controls. For an operator reading this guide: Annotrack shows the market will pay something for the manual annotation path, and that indie supply exists at $9/mo. --- ### Analytics dashboards with no dedicated decision layer **Triple Whale** Price: Free (Founders Dash, $0) · Foundation $219/mo · Automate $749/mo · Enterprise = contact sales ($20M+ annual GMV). Metered on annual GMV — the same product costs more as a seller's revenue grows. Source: triplewhale.com/pricing, 2026-08-07, PRIMARY. Triple Whale is the Shopify DTC analytics subscription built around attribution. Its core is a first-party pixel tracking from ad click to purchase; above it sit creative analytics, a media-mix model for incrementality attribution, profit reporting, LTV forecasting, and Moby — an "Ask Anything" bar with persistent cross-session conversational memory and a History record of past exchanges. The Automate tier adds agentic campaign actions across ad platforms. What Moby's memory contains: conversational context about campaign-performance questions, not a structured record of the seller's own operational decisions with retrospective outcomes attached. Moby IS queryable — that is the correct description — and the distinction from a decision-memory corpus is in what it holds and what it answers. **Honest limit:** Triple Whale is the right call for a seller who needs attribution, creative performance, and cross-channel forecasting. It carries brand recognition in the DTC world, calibrated attribution benchmarks, a support organisation, and a vendor-managed roadmap — advantages that accrue over years to an established subscription business. An operator running Shopgraph will take time to build any of those; state this accurately to every seller who asks. **Lifetimely by AMP** Price: Free (≤50 orders/month) · $79/mo (≤500 orders) · $149/mo (≤3,000 orders) · $299/mo (≤7,000 orders), metered on monthly order volume. Amazon data add-on +$75/mo. Source: useamp.com/pricing, approximately June 2026, SECONDARY. Lifetimely is the profit, LTV, and cohort analytics tool for Shopify and Amazon, now operating as "Lifetimely Profit Agent and LTV by AMP." It derives everything it shows from order and product data — profit by SKU, cohort retention, subscription analytics — without requiring seller input for its core reports. **Honest limit:** Lifetimely is a profit analytics tool, not a decision-memory tool. It will show the revenue trend before and after a price change, but without a named decision record to attribute the change to, the cause remains an inference the seller draws themselves. **Northbeam** Price: Starter from $1,500/mo for brands under $1.5M annual media spend. Professional and Enterprise are contact. Metered on media spend tier and pageview volume. Source: northbeam.io/pricing, 2026-08-07, PRIMARY. Northbeam is server-side multi-touch attribution for brands running significant media budgets across Meta, Google, TikTok, and connected TV. **Honest limit:** The $1,500/mo entry price places Northbeam above the $500K–$10M GMV seller band this guide covers. It appears here because it comes up in category searches and because an operator mapping this space should know where the premium attribution tier begins. It is not a realistic option for the indie seller this market serves. --- ### Own-the-code — the operator-builds-the-platform path **Shopgraph** Price: $149, one-time. Demo: [https://shopgraph.saascode.ai](https://shopgraph.saascode.ai). Purchase: [https://saascode.ai/products/shopgraph](https://saascode.ai/products/shopgraph). Shopgraph is the only entry on this list that an operator purchases once and deploys as a multi-tenant SaaS under their own domain and brand. Each tenant — one Shopify seller — gets their own isolated decision corpus. The Shopify connector ingests campaign launches, listing edits, price changes, and restock events automatically. An outcome-binding cron returns at T+7 and T+30 days, pulls post-event metrics, and retrospectively closes each decision record without seller intervention. Natural-language querying runs over each tenant's corpus under a grounded-refusal contract: empty retrieval returns a structured refusal, not a fabricated answer. The one capability Shopify's own platform leaves to apps — a seller typing their own annotation — is available here alongside the derived capture, for decisions the connector cannot infer from activity data. **Stated cons — the product earns its position in this list only if these are named.** The corpus is near-empty for months zero through three. During that window, a paying seller who asks a question is told, honestly, that the corpus has nothing relevant to answer from. The grounded-refusal contract that makes the product trustworthy at twelve months is also what makes the first weeks uncomfortable. Derived capture is the mitigation: the Shopify connector fills the corpus with what it can infer so the seller's earliest answers may arrive before they have typed a single note. But even a derived corpus needs time to become a meaningful library. Additional constraints: Shopgraph is not an attribution stack, not a forecasting tool, not a profit-analytics dashboard. The Amazon SP-API connector is V1.5 and ships as a roadmap stub in V1.0 — no multi-marketplace querying at launch. Deploying and running the platform is the operator's responsibility: a Supabase project, a Shopify developer app registration, and a hosting environment are required. --- ### Full custom build **Commission it** A custom build delivers exactly the corpus design and query interface the operator specifies. A developer who has already built a Shopify OAuth connector, a pgvector embeddings pipeline with per-tenant isolation, and a multi-tenant natural-language query layer will produce a precisely tailored result. A team encountering each for the first time will discover that each is its own project. Scoping the full stack accurately before committing to this path is worth the time. --- ## How to choose | If the priority is… | The honest recommendation | |---|---| | Zero cost, zero setup, automatic capture | Shopify native — it is on the plan the seller already pays for | | Simplest annotation log for a seller who will use it | Annotrack ($9/mo) — purpose-built, honest price | | Attribution, creative analytics, campaign performance | Triple Whale — right tool, right category | | Profit, LTV, and cohort reporting | Lifetimely — built for those jobs specifically | | High-budget media attribution ($1.5M+ annual spend) | Northbeam — different tier, entry $1,500/mo | | A decision-memory SaaS to deploy for Shopify sellers, code owned by the operator | Shopgraph — the own-the-code option in this list | | Full control over corpus design, willing to scope and build | Custom development | --- ## Honest verdict Shopgraph earns a position in this list in a narrow lane: it is the only entry an operator can purchase once, deploy as a multi-tenant SaaS for Shopify sellers, and operate with a pricing structure that does not impose GMV-based metering on the operator's costs. The tools above it in the capture-method ranking — Shopify's own free layer — cover more of the spectrum automatically, but ship as built-in capabilities within an existing platform, not as a business the operator runs. The adoption risk is real and belongs in every conversation the operator has with prospective tenants before they subscribe: the corpus-building window, the grounded-refusal period in months zero through three, the derived-capture mitigation that is already built into V1 and is what separates this from a tool that waits for a habit to form. An operator who prices annual plans and builds onboarding flows that surface early derived-corpus wins is structurally better positioned than one who pitches the twelve-month library from day one. The verdict is narrow, and it holds. --- ## FAQ **Which analytics tools in this category get more expensive as a Shopify seller's store grows?** Three tools in this market meter pricing on growth metrics. Triple Whale meters on annual GMV — Foundation $219/mo, Automate $749/mo (triplewhale.com/pricing, 2026-08-07, PRIMARY). Northbeam meters on annual media spend and pageview volume, Starter from $1,500/mo (northbeam.io/pricing, 2026-08-07, PRIMARY). Polar Analytics meters on GMV but publishes no prices on its own site. Annotrack charges a flat $9/mo per seller. Shopgraph's tenant plans, as pre-configured in the codebase, are flat — but the operator who deploys it sets the actual pricing and is not bound to the defaults. **What is the difference between Shopify's automatic annotations and a decision-memory platform?** Shopify's automatic annotations mark events the platform detects — product changes, app installs, theme publications — as coloured markers on the analytics chart. Since July 2026, apps can also write markers covering campaigns, discounts, and inventory. What annotations do not contain: the seller's own account of why they acted, context the platform cannot detect, or outcome binding at a configurable interval. Shopgraph stores a structured record of the decision as the seller knows it, binds it to what followed at seven and thirty days, and makes the full history queryable in prose over each tenant's own isolated corpus. **Does Triple Whale's Moby query the same kind of data as Shopgraph's query layer?** Moby is a queryable "Ask Anything" interface with persistent conversational memory — it is correctly described as queryable, and that is not in dispute here. Its corpus contains campaign performance and attribution data. Shopgraph's query layer answers over a structured log of the seller's own operational decisions with retrospective outcomes attached, under a grounded-refusal contract that returns a structured refusal rather than a generated answer when the corpus has nothing relevant. **Is there a Shopify app that lets sellers annotate their own store events without buying a full analytics platform?** Yes. Annotrack (apps.shopify.com/annotrack-annotations) launched May 19, 2026: sellers log any event with a date, category, and note; paid tiers add impact reports and integrations that pull Klaviyo and Mailchimp campaigns automatically. Pricing: Free (≤25 annotations), $9/mo (Growth), $15/mo (Unlimited). Source: apps.shopify.com/annotrack-annotations, 2026-08-07, PRIMARY. Shopgraph is a different purchase category: source code an operator deploys to run a multi-tenant SaaS for many sellers, not a single-seller Shopify app. **What happens to a seller's analytics history if their vendor gets acquired?** A live question in this market: Peel Insights was acquired by Relay Commerce in May 2024; Glew.io was acquired by Everest Group LLC in March 2026 (announced 2026-03-10). Both continue operating, but the entity controlling the seller's data changed. When a corpus lives inside a vendor's infrastructure, an acquisition changes who decides what happens to it. When the corpus lives on infrastructure the operator controls — as with Shopgraph, which runs on the operator's own Supabase instance — an acquisition of the tool vendor changes nothing about the data or the operator's access to it. ## Related reading - [Shopgraph vs Triple Whale](https://saascode.ai/inside/shopgraph-vs-triple-whale.md) - [Road to Shopgraph](https://saascode.ai/inside/road-to-shopgraph.md)